Creating a scatter plot between two variables and a weight using a third variable with colors to distinguish each marker

[Justin Niakamal]

Hello all,

I have tried numerous times to pull this off with out any success. I am interested in creating an employment cluster graphic that plots the growth of an industry against its location quotient. I would like to have those markers weighted on the number of jobs for that industry and each to be colored in order to distinguish them. I have never been able to combine all of these attributes into one graph. So in short, I have three variables; x, y, and z. I want to plot x and y in a scatter plot. I want the markers to be weighted on variable z, and I want all of the markers to be colored differently (to distinguish them from each other). I've attached an image that shows what I'm getting at. Thanks all!

[Oded Mcdossi]

Here is an example of what you are asking for :

sysuse lifeexp, clear
keep if region ==2
seperate lexp , by( country )
tw scatter lexp1-lexp14 gnppc [fw= safewater ] || scatter lexp gnppc, m(none) mlabel(country)  mlabc(black)  xsca(range(35000))  legend(off)

[Justin Niakamal]

That did the the trick. Thank you for your help!